Lithium Linked to Reduced Cancer Risk

Nancy A. Melville

November 15, 2016

Lithium, the gold-standard treatment for bipolar disorder (BD), has been linked to a significantly reduced cancer risk in a dose-dependent manner in a new study.

Investigators at the College of Pharmacy at Kaohsiung Medical University, in Taiwan, found up to a 45% reduction in cancer risk among BD patients who received the highest cumulative doses of lithium. However, some experts question the methodology behind the findings.

"Our findings imply that the lithium use was associated with a lower incidence of overall cancer in patients with BD," first author Yi-Hsin Yang, PhD, told Medscape Medical News.

The study was published in the November issue of the British Journal of Psychiatry.

Potential Mechanisms

Considered the gold standard for treatment of BD, lithium is a known inhibitor of glycogen synthase kinase-3 (GSK-3). The authors note that GSK-3 plays important roles in many diseases, including cancer, immune disorders, metabolic disorders, neurodegenerative diseases, and neuropsychiatric diseases.

However, results of research on the effects of GSK-3 inhibition have been inconsistent, with some studies showing an association with tumor promotion and others showing potential tumor suppression.

To better understand the effect in the treatment of patients with BD, the authors turned to a large population-based database, the National Health Insurance Research Database, in Taiwan. They identified 4729 patients older than 18 years who had BD. Of these, 370 (7.8%) used only lithium, 3250 (68.7%) used only the anticonvulsant valproate, and 1109 (23.5%) used both lithium and the anticonvulsant.

A total of 115 cases of cancer were identified among the patients. In the anticonvulsant-only group, 86 patients (2.65%) were found to have newly diagnosed cancer during follow-up (4.74 cases per 1000 person-years); 29 patients (1.96%) in the combined lithium with or without anticonvulsants group were found to have newly diagnosed cancer (2.66 cases per 1000 person-years).

The incidence rates were higher than those observed in the general Taiwan cancer registry (2.55 to 4.00 per 1000 person-years), which is consistent with previous reports indicating a higher cancer incidence in people with BD.

The median time for follow-up ranged from 5.2 years for the anticonvulsants group to 7.5 years for the lithium and anticonvulsants group. Lithium exposure was associated with a significant reduction in overall cancer risk in comparison with those who used only anticonvulsants (hazard ratio [HR] = 0.735; 95% confidence interval [CI], 0.554 - 0.974).

A cumulative daily defined dose of lithium greater than 215, defined as treatment with a recommended maintenance lithium dose for 215 days or longer, was associated with a 44.8% lower risk for cancer (HR = 0.552; 95% CI, 0.367 - 0.831) in comparison with those who took only anticonvulsants.

An average daily defined dose of 0.90, representative of adherence to medication (810 mg of lithium carbonate or 1187.1 mg of lithium sulfate per day), was associated with an HR of 0.425.

"Bipolar patients who took 0.90 average daily defined dose of lithium would have 57.5% less chance of developing cancer as compared to bipolar patients who took anticonvulsants only," said Dr Yang.

For lithium users, there were trends of decreased cancer risks with respect to all site-specific cancers with the exception of bone, skin, and connective and other soft tissue cancer (HR = 3.012; 95% CI, 0.798 - 11.365) and genitourinary cancer (HR = 1.014; 95% CI, 0.472 - 2.179).

Although noting that the higher risk for cancer of bone, skin, and connective tissue was surprising, Dr Yang cautioned that "given small numbers of cancer cases, future studies will be needed to confirm those findings."

Patients in the lithium group were younger than those in the anticonvulsant-only group (median age, 38.5 vs 45.9). Although the rate of physical comorbidity was lower for these patients, as determined by scores on the Charlson Comorbidity Index (P < .001), the rate of psychiatric comorbidities was higher, with the exception of anxiety.

Because lithium is indicated for the treatment of BD and manic episodes, little is known of the drug's effects relating to cancer outside of that population, Dr Yang said.

The evidence pertaining to those with BD is nevertheless important, he added. Not all epidemiologic studies show increased cancer incidence in the BD population; those that do attribute the increase to various factors.

"Bipolar patients have an unhealthy lifestyle, such as tobacco smoking and alcohol use, which are identified as avoidable risk factors for cancer," said Dr Yang.

"[In addition], BD prompts an immunity-inflammation response, and some inflammation-related marker levels are increased, such as high-sensitivity C-reactive protein and interleukin-1 receptor antagonist, [which] predispose patients to immune-related diseases such as cancer."

Furthermore, "some studies [suggest] that prolactin-raising effects of antipsychotic medication could be related to an increased female-hormone-regulated cancer risk, such as breast cancer, in female schizophrenia."


In a previous study, Anton Pottegård, PhD, associate professor of clinical pharmacology and pharmacy with the University of Southern Denmark, in Odense, and colleagues reported that long-term lithium use was not linked to an increased risk for upper urinary tract cancer, although a smaller study found evidence of an increased risk for renal neoplasia.

Dr Pottegård noted that the new study uses methodology that in many ways is based on speculation.

"Basically, they classify people into groups at the day they fill their first drug, based on whether they at some later point in time used anticonvulsants or not," he told Medscape Medical News.

In addition, the classification of people with respect to "total amount of drug filled" at baseline is problematic, he said.

"Consider an individual starting lithium treatment in 2000. If this person gets a cancer in 2002 and dies 1 year later, he has a markedly lower chance of getting an anticonvulsant and thus being reclassified into the mixed-treatment group than if he had survived until 2009, the last year of the study."

The same applies to an even greater extent when grouping individuals with respect to cumulative use of lithium, he added.

"If you get a cancer and die [and thus] stop treatment, this affects the amount of drug you fill. But as the individuals are classified regarding their total use over the entire period, we will therefore likely [incorrectly] move cancer cases to lower-use stratas, whereas those in the high-use categories are unlikely to have contracted cancer, since they need to avoid the cancer to get into the high use strata.

"This renders the actual interpretation of the study findings very difficult."

Dr Pottegård pointed out that a drug that is seemingly associated with such a strong and general decrease in cancer risk, irrespective of the type of cancer, "contradicts our current understanding of cancer as a very heterogenic disease," he said. "Such a strong effect on all types of cancer is simply not biologically plausible."

A study that was published this year in Bipolar Disorders adds interesting evidence on the issue.

The study included 5442 BD patients in Sweden. Between July 2005 and December 2009, some of these patients were treated with lithium, and some were not. Although the study found no overall increase in cancer in BD patients in comparison with the general population in Sweden, in BD patients who were not treated with lithium, there was a significant increase in the risk for cancer of digestive organs (incidence rate ratio [IRR], 1.47), the respiratory system and intrathoracic organs (IRR, 1.72), and of the endocrine glands and related structures (IRR, 2.60) in comparison with those who were treated with lithium.

"BD was not associated with increased cancer incidence and neither was lithium treatment in these patients," the authors concluded. "Specifically, there was an increased risk of respiratory, gastrointestinal, and endocrine cancer in patients with BD without lithium treatment."

The authors and Dr Pottegård report no relevant financial relationships.

Br J Psychiatry. 2016;209:393-399. Abstract



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.